Estimating relatedness of various data sets is useful in many applications. In social networking, for example, estimating potential interest of a first user in another user may be useful when selecting users to suggest as potential “friends” to the first user, and vice versa. It is also useful in selecting advertising to present (render) to a particular user. To estimate relatedness, a virtual space may be defined so that it describes a given number of dimensions representing different qualities. For example, a virtual reality space may be defined so that the distance between avatars represents similarity of their qualities rather than or in addition to the distance in the “physical” space or distance as it appears on a computer screen or another display. There are other uses for estimating relevance/relatedness as well.
Therefore, it is desirable to facilitate the process of estimating relatedness of real or notional objects, and taking action based on the resulting estimates.